Comparison of <scp>CMIP6</scp> and <scp>CMIP5</scp> models in simulating mean and extreme precipitation over East Africa

Abstract This study examines the improvement in Coupled Model Intercomparison Project Phase Six (CMIP6) models against the predecessor CMIP5 in simulating mean and extreme precipitation over the East Africa region. The study compares the climatology of the precipitation indices simulated by the CMIP models with the CHIRPS data set using robust statistical techniques for 1981–2005. The results display the varying performance of the general circulation models (GCMs) in the simulation of annual and seasonal precipitation climatology over the study domain. CMIP6 multi‐model ensemble mean (hereafter MME) shows improved performance in the local annual mean cycle simulation with a better representation of the rainfall within the two peaks, especially the MAM rainfall relative to their predecessor. Moreover, simulation of extreme indices is well captured in CMIP6 models relative to CMIP5. The CMIP6‐MME performed better than the CMIP5‐MME with lesser biases in simulating Simple Daily Intensity Index (SDII), consecutive dry days (CDD), and very heavy precipitation days &gt;20 mm (R20mm) over East Africa. Remarkably, most CMIP6 models are unable to simulate extremely wet days (R95p). Some CMIP6 models (e.g., NorESM2‐MM and CNRM‐CM6‐1) depict robust performance in reproducing the observed indices across all analyses. OND season shows wet biases for some indices (i.e., R95p, PRCPTOT), except for SDII, CDD, and R20mm in CMIP6 models. Consistent with other studies, the mean ensemble performance for both CMIP5/6 shows better performance as compared with individual models due to the cancellation of some systematic errors in the individual models. Generally, CMIP6 depicts improved performance in the simulation of MAM rainfall compared with CMIP5 models. However, the new model generation is still marred by uncertainty, thereby depicting unsatisfactory performance over the East Africa domain. This calls for further investigation into the sources of persistent systematic biases and a methodology for identifying individual models with robust features that can accurately simulate observed patterns for future usage.